# Parameter Sweep Optimizer

> Spawns 25 parallel parameter sweep configurations, extracts performance curves, plots response surfaces, and designates optimal param configs.

- Skill: `knuckles-team/parameter-sweep-optimizer` (Agent Skill, multi-file: 2 files)
- Install (CLI): `npx skillmds@latest add knuckles-team/parameter-sweep-optimizer`
- Raw SKILL.md: https://api.skillmd.com/api/skills/knuckles-team/parameter-sweep-optimizer/raw
- Safety review: pending
- Works with: Claude Code, Claude.ai, OpenAI Codex
- Category: Coding & Dev Tools
- Author: Knuckles-Team (https://skillmd.com/u/knuckles-team)
- Updated: 2026-09-17
- Page: https://skillmd.com/skills/knuckles-team/parameter-sweep-optimizer

---


# Parameter Sweep Optimizer Workflow

**CONCEPT:EE-011**

Spawns 25 parallel parameter sweep configurations, extracts performance curves, plots response surfaces, and designates optimal param configs.

## Steps

### Step 1: Grid Search Sweeper
**Agent**: `data-fetcher`
**Tools**: `graph_query, sx_search`

Spawns 25 parallel parameter configuration models.
Expected: `multi-parameter-backtest-records`

### Step 2: Metric Extractor [depends_on: grid-search-sweeper]
**Agent**: `compute-engine`
**Tools**: `graph_analyze`

Extracts performance indicators and fit scores for each sweep node.
Expected: `extracted-response-metrics`

### Step 3: Surface Plotter [depends_on: metric-extractor]
**Agent**: `risk-assessor`
**Tools**: `graph_query, graph_analyze`

Generates a 3D parameter response surface map to identify stable basins.
Expected: `volatility-response-surface-plot`

### Step 4: Config Selector [depends_on: surface-plotter]
**Agent**: `report-generator`
**Tools**: `graph_write, document_tools`

Selects the parameters that minimize overfitting and registers them.
Expected: `finalized-parameter-configs`

### Step 5: KG Persistence [depends_on: config-selector]
**Agent**: `report-generator`
**Tools**: `graph_write`

Persist workflow results as nodes and edges in the Knowledge Graph.
Create appropriate typed nodes with metadata and link to existing domain entities.

## Output
- Parameter Sweep Optimizer results persisted in KG
- Structured report (MD/PDF)
- Audit trail with timestamps and agent attributions

## Execution

Run this workflow as a dependency-ordered DAG. Steps with no unmet `depends_on` run in parallel; dependents run after their prerequisites complete.

- **Run first (in parallel):** Step 1 — Grid Search Sweeper
- **After level 0:** Step 2 — Metric Extractor
- **After level 1:** Step 3 — Surface Plotter
- **After level 2:** Step 4 — Config Selector
- **After level 3:** Step 5 — KG Persistence

**Execution:** If graph-os is reachable, offload the whole DAG via `graph_orchestrate action=execute_workflow` (or the `kg-delegate` skill) for true parallel/swarm execution. Otherwise execute the steps natively in dependency order: run steps with no unmet `depends_on` in parallel, then their dependents.